Re: Single-elim

• From: stipenn
• Date: Wed, 01 Aug 2001 12:22:11 -0700
• Message-ID: <000TMGLN3.000J22KUD.12170551-quizbowl_at_...>

Bill wrote:

<<I guess what I'm
really advocating here is . . . something that doesn't
eliminate a player from the playoff based on a single
round. We've heard an endless chorus against single-elim
playoffs, so I guess my . . . comments are just a variant
on the refrain.>>

Although, as others
have pointed out, it is not single-elim _per se_ that
is the problem, but the fact that single-elim can
magnify defects caused by unusual factors, whatever they
may be. (e.g., The quarterfinal packet at Beaver
Bonspiel 4 that had 9 TUs between science and classical
music, and the octofinals packet at Penn Bowl
9.)

On a side note: I don't know if this would work, but
an idea I had--in an attempt to avoid both seeding
and vast differences in room strength--would be a
random draw for initial placements. The catch is that,
once picked, each player would pick the room in which
s/he would play. My thinking is that players would
probably sort themselves automatically into rooms of
roughly equal strength (or, at least, more equal than a
truly random draw would yield).

Of course, to
continue the old refrain, if anyone can come up with a
good, fair way to eliminate single-elim at Penn Bowl,
I'm all ears. [There are a few caveats: (1) it must
not require more than 21 rounds total; (2) every team
must get at least 14 matches; (3) teams must not be
eliminated(*) before playing 10 matches; (4) it must not
require more than 6 house packets (and preferably fewer);
and (5) it must be simple enough that someone not
mathematically inclined could run it.]

(*)I am referring
to the division of the field into "championship" and
"consolation" brackets; obviously, a team that loses their
first six matches is probably out of the running for
any sort of playoff.



--STI